View Full Version : Work flows for more ambitious projects?
02-19-2010, 04:26 AM
I am looking to work on an environment for my own personal portfolio and by my estimates it will be a couple of months work of work. I have never done anything as ambitious or as tim intensive on my own so i was wondering if i could get some feed back on how i am planning out the project and the work flow i intend to take.
The piece will be consist of 10 modular pieces,maybe 11. There will be a high poly,low poly,zbrush and texture pass per piece. The pieces will have alot of decorative elements so in other words,alot of work. I have tried projects less ambitious in the past but ive always jumped right into them with not alot of planning and it had killed the momentum. So heres the workflow im thinking.
1.references pics (done)
2. Indentify how many modular pieces there will be (done)
3.get the proportions to each pieces (done)
4.make a really rough block out of the mesh for each piece
now here is where it can get complicated.Because these elements are supposed to work as a whole.
6. - I dont know if its best to make all of the high polys for each building,then the lows,unwrap,zbrush highs and make normals.
- or if its best to take this process one piece at a time,if i have a good and measured out block out the pieces will fit but i worry about the overall look in the end.
- or make the lows first to get an over idea of the building,highs,z brush,unwrap and bake normals.
Which one of these paths seemed better suited for the scal of the project?
7. Texturing will be the next pass for all of them.
8.Get em in U3,set up a scene
9.rejoice at my epic sized project.
I want to be organized enough to make a breathtaking portfolio piece that shows the full depth of my ability.
i'd go with the third option for 6 because you might always figure out ways to save UV space or sculpting time if you do your lowpoly first. that's at least what happens to me a lot :D
02-19-2010, 08:34 AM
I'm currently working on a fairly large scale project as well, I figure I'm about 2/3rds finished maybe 1-2 months more to go, hopefully...
From my personal experience and mistakes along the way, I'd say along with the things you've already completed:
1. Do any technology tests you need to do before starting in on the real production so that you are sure the things you plan to do will work without having to worry about them until you get to them.
It might even be a good idea to take one of the smaller pieces and make it from start to finish. Then you can do the rest in stages with the knowledge that everything should work out just fine.
I really should've done this, but so far I've just been figuring out things as I go along and It's kind of stressful.
2. Block everything out in UDK, make sure the pieces fit and the scene composition is worth doing.
A big mistake I made early on was constructing my blockout scene in Modo then having to reconstruct it in Unreal. I should've known better, oh well live and learn.
3. Make the high poly models from easiest to hardest, to keep consistency with each other. You'll get better as you go along so the hard ones will be easier when you get to them.
I think it takes some serious self control and patience to actually follow through with this step before moving on. I failed at this because I was anxious to test things out and to see something near done in engine so I started to make low poly models, UV, then bake them before finishing all my high poly models.
4. Model the low poly models.
5. Import the low poly models into UDK replacing the blockout models to double check that they still work together.
6. Unwrap UVs and bake textures.
7. Double check baked maps in UDK, and check lightmap baking.
8. Create remaining textures.
9. Add anything necessary to complete the scene, decals, sky domes, etc...
I personally think that I should have made a few smaller projects rather than one really large one. Sometimes it gets boring to work on the same thing all the time, and I've been sidetracked more than once because of it.
In the end I believe you are more experienced than I am with these kinds of things so take from it what you will. Just figured I'd share my thoughts on the subject.
Good luck :)
02-20-2010, 09:45 AM
Excellent,alot of good advice.
02-20-2010, 12:17 PM
Organization is key. You should be alright as long as you do the simple things, things that newbies tend to forget. Texturing is the time hog for me, personally. Making sure I have as many layers as I need to make quick and easy changes in photoshop to my textures is key. There are always little things you should do in all the programs you use that speed up time, or save time when you have to make changes. In my experience, you're gonna have to make changes to your environment no matter what, it's just how projects evolve and get better. The real key is to plan things out as much as possible before you start on production. The more of an idea you have where overall layout, art direction, etc. is going, the less time you'll spend reworking stuff due to poor planning. Everyone has their own workflow for things, and they all have their flaws and perks. The key is to keep working at it, and project by project, you'll develop something that works for you. Practice makes perfect.
Sorry, that was a bit jumbled up. Take from it what you will.
02-23-2010, 01:39 AM
Thanks again for the feedback,last night i began with this little(huge) project, Im going to construct all of the low poly environments first then do the highs. I think it will be more beneficial to see all of the lows in the UDK and see how they all work as a whole and when i am satisfied ill begin making highs then unwraps and bakes. A good reason i see for making lows first is because ill be able to see such things as what pieces will benefit from mirroring,wha will be visible in the editor and like that i can determine what area will need more details than others. Fixing lows is less time concuming as well.
3dRyan,i agree about organization,i think its the modeling and z brush that will take the most time and paitience. What i am making is an old building i have seen in Berlin that is before the war which is bullet riddled,destroyed and highly detailed. An environment artists dream. The buuilding is mostly the same material so with a good bas texture that is well organized i can take the base and use it in every texture which will be a nice time save. The little details will be the main time sink workwise.
Does 2048 texture seem a little decadent for 11 elements? Maybe i dont need the resolution but then again i am trying to make a wow piece that shows fully what i can do talent wise.
02-23-2010, 03:42 AM
Productivity in 11 Words
One thing at a time. Most important thing first. Start now.source (http://twitter.com/scottros/statuses/8331751617)
02-23-2010, 08:05 AM
@render, stared my reoports because of that sentence^^
productifity makes one feel awesome
vBulletin® v3.8.4, Copyright ©2000-2013, Jelsoft Enterprises Ltd.